[en] Van Flandern has postulated a variation of the gravitational constant at the rate approximately -8 x 10^-11/yr. This variation, consistent with Hoyle-Narlikar and Dirac cosmologies, has been assumed in the computation of a 5 x 10⁹ yr theoretical isochrone. Present results show that, even for this age, theory predicts a cluster turn-off luminosity approximately 0.5 to 1.0 mag fainter than the observed turn-offs of globular clusters. Unsatisfactory agreement between theoretical and observed luminosity functions is also indicated. (author)